Abdul Latif Ahmad for his wonderful supervision and his unrelenting support, expert guidance, valuable comments and the enormous time and effort he rendered throughout my research work

MIXED MATRIX VANADIUM OXIDE CATALYTIC NANOCOMPOSITE MEMBRANE FOR STYRENE OXIDATION

by

BEHNAM KOOHESTANI

Thesis submitted in fulfillment of the requirements for the degree of

Doctor of Philosophy

March 2012

ii

ACKNOWLEDGMENTS

First, I would like to thank the Almighty God for His infinite mercy and protection upon the accomplishment of my studies. I would like to express my genuine gratitude to my supervisor, Prof. Dr. Abdul Latif Ahmad for his wonderful supervision and his unrelenting support, expert guidance, valuable comments and the enormous time and effort he rendered throughout my research work. I would not have better supervision and feel to be a lucky person to work with him. I would also like to extend my heartfelt thanks to Dr. Ooi Boon Seng and Prof. Dr. Subhash Bhatia for their brilliant comments, encouragement and for providing me continuous advice throughout my studies. I really was honored to have the opportunity to work under the supervision of all of them.

I would also like to express my appreciation to the Dean, Prof. Dr. Azlina BT.

Harun @ Kamaruddin and Assoc. Prof. Dr. Lee Keat Teong, Assoc. Prof. Dr.

Mohamad Zailani Bin Abu Bakar and Assoc. Prof. Dr. Ahmad Zuhairi Abdullah, Deputy Deans of the School of Chemical Engineering USM, for their continuous support and help rendered throughout my studies. My sincere thanks go to all the respective lecturers, staff and technicians in the School of Chemical Engineering for their cooperation and support especially Prof. Dr. Bassim H. Hameed, Dr. Mohd Azmier Ahmad, Dr. Tan Soon Huat , Dr. Lim Jit Kang , Dr. Suzylawati Binti Ismail, Dr. Leo Choe Peng and my best friend Dr. Low Siew Chun. Great appreciation goes to the former Deputy Deans, Dr. Syamsul Abdul Syukor and Dr. Zinal Ahmad.

Thousand thanks to Mr. Shamsul Hidayat Shaharan Mr. Mohd Roqib Rashidi, Mr.

Muhd Arif Mat Husin, Mr. Mohd. Faiza Ismail and Mrs. Latifah Abd. Latif for their valuable and kind help in laboratory works. I highly appreciate all the help and supports spared from Mrs. Aniza Abd. Ghani , Mrs. Normie Hana A. Rahim, Badilah

iii

Baharom, Hasnah Hassan, Sharida Sajili and Fatimah BT. Jahan Khir in smoothing my studies through their assistance in the official work throughout my studies. I am also indebted to School of Industrial Technology and School of Material and Mineral Engineering in USM for their assistance in some of my analysis work in this study.

The financial supports provided by MOSTI (Project No. 03-01-05-SF0401) And RU grants provided by Universiti Sains Malaysia are gratefully acknowledged. I would like to extend my sincere and deepest gratitude to all my adored friends, in Malaysia and in Iran for their unparalleled help, kindness and moral support towards me.

Thank you for always being there for me. I hope we all have a very bright future undertaking ahead. Very special thanks goes to my dear friends Mr. Muhammad Azan , Mr. Hafiz Faisal, Babak, Farhad, Hamed, Farshid, Zahra, fatemeh, Mr.

Senthil Kumar Mr. Ali Sabri, Thiam Leng, Pei Ching, Ee Mee and Moses A.

Olutoye for their useful help and companionship in the lab. Also do wish to express my deepest appreciation to Universiti Sains Malaysia for providing me a warm environment to feel at home.

Last but definitely not least, my deepest and most heart-felt gratitude to my beloved mum, and my adored dad, for their endless love and support. I need to thank very especially to my darling wife and sweetheart son for all those innumerable things I could not possibly have done with them. I would like to dedicate this PhD thesis to them. To my wonderful sisters and my kind brother, my parents my brother and sister in low for their love and care. To who are directly and indirectly involved in this research, your contribution given shall not be forgotten. My appreciation goes to all of you.

Sincerely

Behnam Koohestani

iv

TABLE OF CONTENTS

ACKNOWLEDGMENT ii

iv TABLE OF CONTENT

LIST OF TABLE xiii LIST OF FIGURES xvi

xxi LIST OF PLATES

xxiii LIST OF ABBREVIATION

xxv LIST OF SYMBOLS

xxvi ABSTRAK

xxviii ABSTRACT

1 CHAPTER 1 - INTRODUCTION

Membrane Technology 1 1.1

2 Application of Catalyst to Homogeneous and Heterogeneous Reaction

1.2

4 Nanoparticles Catalysis

1.3

4 Common Nanoparticle Supports

1.4

5 Metal Oxide Supports

1.4.1

5 Carbonaceous Supports

1.4.2

6 Polymer Supports

1.4.3

6 Porous membranes as catalyst supports

1.4.4

7 Membranes Reactors

1.5

8 Retainment of Homogeneous Catalysts

1.5.1

9 Membrane Reactors for the Preferential Removal of a Species

(Membrane Extractors) 1.5.2

10 Membrane Reactor for Distributed Addition of a Reactant (Membrane

Distributors) 1.5.3

v

11 Membrane Reactors for Control of Reactant Contact (Membrane

Contactors) 1.5.4

12 Interfacial Contactors

1.5.4(a)

13 Flow-Through Contactors

1.5.4(b)

15 Benzaldehyde synthesis and Application

1.6

16 Problem Statement

1.7

20 Research Objectives

1.8

20 Scope of the Study

1.9

20 Synthesis and Characterization of Nanocomposite Catalyst

1.9.1

21 Synthesis and characterization of Mixed Matrix Vanadium oxide

Catalytic Nanocomposite Membrane 1.9.2

21 Design and Fabrication of Liquid Permeation membrane Test Rig

1.9.3

21 Kinetic study

1.9.4

22 Organization of the Thesis

1.10

CHAPTER 2 - LITERATURE REVIEW 24

24 Hydrocarbons

2.1

25 Selective Oxidation of Hydrocarbons

2.2

28 Hydrogen Peroxide

2.3

29 Oxidation of Hydrocarbon with Hydrogen Peroxide

2.4

29 Nanotechnology and Nanostructured Materials by Template Synthesis

2.5

30 Different Template Type Used for Catalyst Preparation

2.6

30 Track-Etch Membrane

2.6.1

30 Porous Anodic Aluminum Oxide (AAO) Membrane

2.6.2

vi

31 Porous Polymeric Materials

2.6.3

31 Deposition Strategies for Preparing Nanostructured Materials

2.7

32 Sol-Gel Deposition

2.7.1

Chemical Vapor Deposition (CVD) 33 2.7.2

Atomic and Physical Properties of Vanadium 34 2.8

Nano-Scale Vanadium Oxides 35 2.8.1

Hydrothermal synthesis of vanadium oxide nanotube 36 2.8.1(a)

The Importance of Vanadium in Heterogeneous Catalysis 38 2.9

The Importance of Different Catalysts in Styrene Oxidation and 40 the benchmarking with other catalyst

2.10

The Importance of Vanadium in Styrene Oxidation 41 2.11

Catalytic Membrane in Oxidation and Epoxidation 42 2.12

Catalytic Membrane in Styrene Oxidation 48 2.13

Functionalization of Carbon Nanotubes 50 2.14

Wet Chemistry Method of Filling Nanotubes 51 2.15

Summary 52 2.16

53 CHAPTER 3 - MATERIALS AND METHODS

Introduction 53 3.1

Materials and Chemicals 53 3.2

Synthesis of Catalysts and Membrane 53 3.3

Synthesies of Catalyst Nanoparticles and Reaction Procedure 55 3.3.1

Synthesis of Vanadium Oxide Nanotubes Via an Ultrasonic 55 Method

3.3.1(a)

3UHSDUDWLRQRI-Al2O3Nanoparticles 57 3.3.1(b)

vii Preparation of Catalyst V_x2\-Al₂O₃ 58 3.3.1(c)

Reaction Procedure for the Oxidation of Styrene with VxOy- 58 Al2O3Catalyst

3.3.1(d)

7B 60

Catalytic Activity Determination for the Oxidation of Styrene with VxOy-Al2O3 Catalyst with the Adsorption of Reaction Water by Anhtdrous MgSO4

3.3.1(e)

Pretreatment of MWCNTs 60 3.3.1(f)

Preparation of VxOy- CNTs Catalysts 61 3.3.1(g)

Preparation of VxOy-CNTs-&RDWHGE\-Al2O3Catalysts 61 3.3.1(h)

Preparation of Anhydrous Hydrogen Peroxide 62 3.3.1(i)

Preparation of Membrane and Reaction Procedure 62 3.3.2

Preparation of Alumina Pellet Support 62 3.3.2(a)

Preparation of AlOOH Sol (Boehmite sol) 63 3.3.2(b)

Preparation of AlOOH Sol with VxOywCNTs- Catalysts by 63 Addition of PVA Following Peptization

3.3.2(c)

Preparation of AlOOH Sol with VxOy-CNTs Catalysts by 64 Addition of PVA or PVA+PEG Prior to Peptization 3.3.2(d)

Preparation of AlOOH Sol with VxOy-NTs Catalysts by 64 Addition of Mixture of PVA/PEG Prior to Peptization 3.3.2(e)

Preparation of AlOOH Sol with VxOyCatalysts by Additi.on of 65 Mixture of PVA/PEG Prior to Peptization

3.3.2(f)

Coating of Mixed Matrix Catalytic Membrane 65 3.3.2(g)

Styrene Oxidation: Experimental Set-Up and Analysis 66 3.3.2(h)

Liquid permeation Membrane Test Rig Setup 66 3.4

Setup Instruction 68 3.4.1

Reactor Assembly 69 3.4.2

Process Study Using Liquid Permeation Membrane Test Rig 70 3.5

viii Operating Steps 73

3.5.1

Product Analysis 73 3.6

Characterization of Catalyst and Membrane 74 3.7

X-Ray Diffraction (XRD) Analysis 75 3.7.1

Scanning Electron Microscopy (SEM) and Energy Dispersive 76 X-ray Microanalysis (EDX)

3.7.2

Transmission Electron Microscopy (TEM) 76 3.7.3

Thermal Gravimetric Analysis (TGA) 77 3.7.4

Fourier Transformed Infrared (FT-IR) Spectroscopy 77 3.7.5

Nitrogen Adsorption-Desorption Measurement 77 3.7.6

Zeta Potential and Particle Size of Al₂O₃ 78 3.7.7

Statistical Analysis of Data 78 3.8

Non-linear Mode 79 3.8.1

Design of Experiments (DOE) 81 3.8.2

Kinetic Study of Styrene Oxidation 82 3.9

86 CHAPTER 4 - RESULTS AND DISCUSSION

Preparation & Characterization and Reaction of Catalysts 86 4.1

Vanadium Oxide Nanotubes (VxOy- NTs) Catalyst 86 4.1.1

Effect of Ultrasonic 86 4.1.1(a)

Scanning Electron Microscopy (SEM) 88 4.1.1(b)

Transmission Electron Microscopy (TEM) 90 4.1.1(c)

X-Ray Diffraction (X-ray) 92 4.1.1(d)

Fourier Transformed Infrared (FT-IR) Spectroscopy 92 4.1.1(e)

ix

Thermal Gravimetric Analysis (TGA) for Checking Thermal 93 Stability

4.1.1(f)

Nitrogen Adsorption-Desorption Measurement 94 4.1.1(g)

Stability Test on Vanadium Oxide Nanotubes (VxOy- NTs) 95 Catalyst

4.1.1(h)

VxOy-Al2O3Nanoparticles Catalyst 96 4.1.2

Nitrogen Adsorption-Desorption Measurement 96 4.1.2(a)

Transmission Electron Microscopy (TEM) 98 4.1.2(b)

X-Ray Diffraction (XRD) 98 4.1.2(c)

Energy Dispersive X-ray Microanalysis (EDX) 99 4.1.2(d)

Application of VxOy-Al2O3Catalyst for styre oxidation 100 4.2.1

Model fitting and ANOVA 100 4.2.1(a)

Effect of Catalyst Concentration on Styrene Conversion 104 4.2.1(b)

Effect of Various Reaction Times on Styrene Conversion 105 4.2.1(c)

Effect of H2O2Concentration on Styrene Conversion 105 4.2.1(d)

Effect of Different Reaction Temperatures on Styrene 110 Conversion

4.2.1(e)

Optimization of styrene oxidation 114 4.2.1(f)

The Effect of Water on the Oxidation Reaction 114 4.2.1(g)

Solvent Effect 115 4.2.1(h)

Oxidation of Styrene by Anhydrous Hydrogen Peroxide on 116 VxOy/-Al2O3Nanoparticle Catalysts with water removal 4.2.2

Reaction Procedure for the Oxidation of Styrene 116 4.2.2(a)

X-Ray Diffraction (XRD) 116 4.2.2(b)

Design of Experiment (DOE), Model Fitting and ANOVA 117 4.2.2(c)

x Effect of Catalyst Concentration 121 4.2.2(d)

Effect of Reaction Time 121 4.2.2(e)

Effect of the H2O2/Styrene Ratio on Oxidation 125 4.2.2(f)

Effect of Temperature 125 4.2.2(g)

Optimization of styrene oxidation 129 4.2.2(h)

Solvent Effect 129 4.2.2(k)

Effect of Water Scavenging on Oxidation 132 4.2.2(l)

Vanadium Oxide (VxOy) Supported on Multi-Walled Carbon 133 Nanotubes (CNTs) (VxOy- CNTsDQG-Al2O3Coated VxOy- CNTs

4.3

Catalyst Structure (TEM) 133 4.3.1

BET Surface Area 134 4.3.2

Infrared Spectroscopy (FT-IR) 135 4.3.3

Thermo Gravimetric Analysis (TGA) 138 4.3.4

Catalytic Oxidation 139 4.3.5

Effect of the Composite Catalysts on Styrene Oxidation 139 4.3.5(a)

Effect of Reaction Time 140 4.3.5(b)

Effect of Reaction Temperature 141 4.3.5(c)

Effect of H2O2Concentration 142 4.3.5(d)

Effect of Solvent 142 4.3.5(e)

Optimization of Reaction Conditions 143 4.3.5(f)

Characterization and Reaction of Catalytic Membrane 145 4.4

Mixed Matrix Vanadium Oxide Catalytic Nanocomposite 145 (V_xO_y-CNTs) Membrane for Styrene Oxidation

4.4.1

xi

Effect of Ultrasonic Treatment on Coating of Membrane 145 4.4.1(a)

Zeta Potential and Particle Size of Al2O3 147 4.4.1(b)

Nitrogen Adsorption-Desorption Measurement 148 4.4.1(c)

Transmission Electron Microscopy (TEM) 150 4.4.1(d)

Scanning Electron Microscopy (SEM) 152 4.4.1(e)

Effect of Concentration of Al2O3Sol and Binder 153 4.4.1(f)

Thermal Gravimetric Analysis (TGA) 156 4.4.1(g)

X-Ray Diffraction (XRD) 157 4.4.1(h)

Styrene Oxidation using Catalytic Membrane 158 4.4.2

Effect of Coating Time on Flux and Conversion 158 4.4.2(a)

Styrene oxidation: experimental set-up and analysis 161 4.4.2(b)

Design of Experiment (DOE) 161 4.4.2(c)

Effect of catalyst loading 163 4.4.2(d)

Effect of differential pressure 163 4.4.2(e)

Effect of the H2O2/styrene ratio on oxidation 167 4.4.2(f)

Effect of temperature 170 4.4.2(g)

Characterization and comparison of Mixed Matrix Vanadium 175 Oxide Catalytic Nanocomposite (V_xO_y-CNTs, V_xO_y-NTs and VxOy/Al2O3nano particle) Membrane for Styrene Oxidation 4.4.3

Nitrogen Adsorption-Desorption Measurement for Different 176 Type of Membranes

4.4.3(a)

Scanning Electron Microscopy (SEM) 178 4.4.3(b)

Atomic Force Microscopic (AFM) 180 4.4.3(c)

Energy Dispersive X-ray Microanalysis (EDX) 181 4.4.3(d)

xii X-ray diffraction (XRD) 184 4.4.3(e)

Styrene oxidation comparison study with mixed matrix catalytic 185 membrane prepared with different shape of catalyst

4.4.3(f)

Kinetic Study of Styrene Oxidation 186 4.5

Kinetic Study of Styrene Oxidation with nanoparticle catalyst 186 4.5.1

Kinetic Study of Styrene Oxidation with mixed matrix catalytic 192 membrane

4.5.2

197 CHAPTER 5 - CONCLUSIONS AND RECOMMENDATIONS

197 5.1 Conclusion

200 5.2 Recommendation

REFRENCES 202 APPENDIXES 231

LIST OF PUBLICATIONS 235

xiii

LIST OF TABLES

Page Table 1.1 Operating limits and restrictions for various membrane

materials

8 Table 2.1 Industrial catalytic processes using vanadium oxides 39 Table 2.2 Conversion of styrene, product selectivity with different

catalyst

43

Table 2.3 Conversion of styrene, product selectivity with vanadium base catalyst

46

Table 2.4 Catalytic membrane used for oxidation or epoxidation reaction

49

Table 2.5 Catalytic membrane used for styrene oxidation reaction 50 Table 3.1 List of materials and chemicals used in this study 54

Table 3.2 List of equipment used 55

Table 3.3 list of main components of the test rig 68

Table 3.4 Retention time for styrene oxidation of the GC 74 Table 3.5 Effective factors, upper and lower limit for styrene oxidation

reaction

82 Table 4.1 Comparison of significant model parameters for styrene

conversion and benzaldehyde selectivity

101

Table 4.2 The experimental data for the oxidation of styrene 103

Table 4.3 ANOVA table for styrene oxidation 104

Table 4.4 Limitations adjusted to reach the optimal value for styrene oxidation to benzaldehyde

114

Table 4.5 Styrene oxidation employing aqueous H2O2 (30%) and anhydrous H2O2

115 Table 4.6 Styrene oxidation employing various solvents 116

xiv

Table 4.7 The experimental data for styrene oxidation 119

Table 4.8 ANOVA table for styrene oxidation 120

Table 4.9 Limitations adjusted to reach the optimum value for styrene oxidation to benzaldehyde

129

Table 4.10 Styrene oxidation with various solvents base on selectivity % 131 Table 4.11 Styrene oxidation employing anhydrous H2O2 and the

adsorption of reaction water with MgSo4base on yield%

132 Table 4.12 The surface area, pore diameter and pore volume of raw

materials and catalysts

135

Table 4.13 Styrene oxidation employing various solvents 144 Table 4.14 Zeta potential and mean particle size sol ALOOH 148 Table 4.15 The surface area and pore volume of raw materials and

catalysts

149

Table 4.16 Effect of binder and sol concentration on membrane appearance

155 Table 4.17 Amount of flux, conversion, benzaldehyde and styrene oxide

selectivity with different time of coating by alumina sol include VxOy- CNTs catalyst

161

Table 4.18 Independent variables and their coded and actual values used in the response surface study

162 Table 4.19 The experimental data for the oxidation of styrene 168

Table 4.20 ANOVA table for styrene oxidation 169

Table 4.21 Limitations adjusted to reach the optimal value for styrene oxidation to benzaldehyde

175

Table 4.22 The surface area and pore volume of raw materials and catalysts

177 Table4.23 Compared Mixed matrix catalytic membrane prepared from

VxOy/CNT, VxOy-NTs and VxOy-Al2O3 nano particle for styrene oxidation and related flux with Differential pressure SEHWZHHQIHHGVLGHDQGSHUPHDWHVLGH

186

Table 4.24 Amount of styrene conversion and selectivity at different catalyst and different time

190

xv

Table 4.25 Amount of specific rate constant and activation energy for for benzaldehyde and styrene oxide

191 Table 4.26 Amount of styrene conversion, flux and selectivity at

different differential pressure

195

Table 4.27 Amount of specific rate constant and activation energy for for benzaldehyde and styrene oxide on mixed matrix membrane

196

xvi

LIST OF FIGURES

page

Figure 1.1 Effect of catalyst on a chemical reaction 3

Figure 1.2 Retainment of homogeneous catalysts 9

Figure 1.3 Extractor Membrane 10

Figure 1.4 Membrane reactors for distributed addition of a reactant 11 Figure 1.5 Membrane reactors for control of reactant contact 12 Figure 1.6 Schematic diagram showing the flow of reactants through a

membrane operated as a (a) flow-through contactor, and around a catalyst bead from a (b) traditional fixed bed reactor

14

Figure 2.1 Important industrial organic chemicals produced by heterogeneous oxidation

26

Figure 2.2 Diagram for partial and complete oxidation reaction of hydrocarbon VKRZLQJ WKH SURILOHV IRU HQWKDOSLHV RI WRWDO R[LGDWLRQ +TOx and SDUWLDO R[LGDWLRQ +Pox with the corresponding activation energy profiles for complete oxidation, EaTOxand partial oxidation, EaPOx

27

Figure2.3 Coordination polyhedra: (a) Tetrahedron; (b) Square pyramid; (c) Octahedron ; (d) Triangular pyramid

35 Figure 2.4 Functionalization possibilities for SWNTs: A) defect-group

functionalization, B) covalent sidewall functionalization, C) noncovalent functionalization with surfactants, D) noncovalent exohedral functionalization with polymers, and E) endohedral functionalization

51

Figure 3.1 Experimental work flowchart 56

Figure 3.2 Reaction of styrene oxidation 59

Figure 3.3 Schematic diagram of liquid permeation membrane test rig 71 Figure 3.4 Choromatogram of styrene and mixture of product 75 Figure 4.1 Schematic presentation of one layer formation VxOywhexadecylamine 91

xvii

Figure 4.2 XRD pattern of VxOywhexadecylamine duration of hydrothermal 5 days

92

Figure4.3 FT-IR spectra of VxOy-NTs 93

Figure 4.4 TGA of VxOy-NTs produced by 5 days hydrothermal treatment 94 Figure 4.5 Adsorption/desorption isotherms of VxOy-NTs(5 days hydrothermal

treatment)

95

Figure 4.6 Adsorption/desorptionLVRWKHUPVRID-Al2O3(b) VxOy/-Al2O3 97 Figure 4.7 Pore size distribution of (a)-Al2O3(b) VxOy/-Al2O3 98 Figure 4.8 XRD patterns of the V2O5, Al2O3and catalyst 99 Figure 4.9 EDX analysis of VxOy-Al2O3catalyst 100 Figure 4.10 Response surface plot showing the effect of catalyst amount and time

and their mutual effect on (a) styrene conversion (b) interaction graph (c) predicted vs. actual

107

Figure 4.11 Response surface plot showing the effect of catalyst amount and time and their mutual effect on (a) benzaldehyde selectivity (b) Interaction Graph (c) predicted vs. actual

108

Figure 4.12 Response surface plot showing the effect of catalyst amount and time and their mutual effect on (a) styrene oxide selectivity (b) Interaction Graph (c) Predicted vs. actual

109

Figure 4.13 Response surface plot showing the effect of oxidant and temperature and their mutual effect on (a) styrene conversion (b) Interaction Graph (c) predicted vs. actual

111

Figure 4.14 Response surface plot showing the effect of oxidant and temperature and their mutual effect on (a) benzaldehyde selectivity (b) Interaction Graph (c) predicted vs. actual

112

Figure 4.15 Response surface plot showing the effect of oxidant and temperature and their mutual effect on (a) styrene oxide selectivity (b) Interaction Graph (c) predicted vs. actual

113

Figure 4.16 XRD patterns of the V2O5, Al2O3, fresh catalyst and catalyst after 117

xviii reaction

Figure0 4.17 Response surface plot showing the effect of catalyst and time and their mutual effect on (a) styrene conversion (b) Interaction Graph (c) predicted vs. actual

122

Figure 4.18 Response surface plot showing the effect of catalyst amount and time and their mutual effect on (a) benzaldehyde selectivity (b) Interaction Graph (c) predicted vs. actual

123

Figure 4.19 Response surface plot showing the effect of catalyst amount and time and their mutual effect on(a) styrene oxide selectivity (b) Interaction Graph (c) predicted vs. actual

124

Figure 4.20 Response surface plot showing the effect of oxidant and temperature and their mutual effect on (a) styrene conversion (b) Interaction Graph (c) predicted vs. actual

126

Figure 4.21 Response surface plot showing the effect of oxidant concentration and temperature and their mutual effect on (a) benzaldehyde selectivity (b) Interaction Graph (c) predicted vs. actual

127

Figure 4.22 Response surface plot showing the effect of oxidant and temperature and their mutual effect on (a) styrene oxide selectivity (b) Interaction Graph (c) Predicted Vs. Actual

128

Figure 4.23 FT-IR spectra of the catalysts: (a) VxOy-CNT /-Al2O3, (b) VxOy- CNT, and (c) VxOy-Al2O3

137

Figure 4.24 TGA of the catalysts from room temperature to 900¡C 138 Figure 4.25 Effect of catalyst concentration on styrene oxidation (reaction

temperature of 60C, reaction time of 4.5 h, and molar ratio of H2O2/styrene of 2:1)

140

Figure 4.26 Effect of time on styrene oxidation (reaction temperature of 60C, catalyst loading of 0.014 g/ml for VxOy-Al2O3and 0.008 g/ml for VxOy-CNT and VxOy-CNT/Al2O3, and molar ratio of H2O2/styrene of 2:1)

141

Figure 4.27 Effect of temperature on styrene oxidation (reaction time of 5 h, catalyst loading of 0.014 g/ml for VxOy-Al2O3 and 0.008 g/ml for VxOy-CNT and VxOy-CNT/-Al2O3, and molar ratio of H2O2/styrene of 2:1)

141

Figure 4.28 Effect of concentration of oxidant on styrene oxidation (reaction time of 5 h, catalyst loading of 0.014 g/ml for VxOy- Al2O3and 0.008 g/ml for VxOy-CNT and VxOy-CNT/Al2O3, and reaction temperature of 60C)

142

xix

Figure 4.29 pore size distribution of unsupported membrane 150 Figure 4.30 Adsorption isotherms of unsupported sample 151 Figure 4.31 Weight loss curve for catalyst and unsupported membrane 157 Figure 4.32 XRD patterns of unsupported membrane with different condition of

preparation

158

Figure 4.33 Response surface plot showing the effect of catalyst loading and differential pressure (delta p), and their mutual effect on the (a) styrene conversion (b) Interaction Graph (c) Predicted vs. actual

164

Figure 4.34 Response surface plot showing the effect of catalyst loading and differential pressure (delta p), and their mutual effect on the(a) benzaldehyde selectivity (b) Interaction Graph (c) Predicted vs.

actual

165

Figure 4.35 Response surface plot showing the effect of catalyst loading and differential pressure (delta p), and their mutual effect on the (a) styrene oxide selectivity (b) Interaction Graph (c) Predicted vs. actual

166

Figure 4.36 Response surface plot showing the effect of oxidant and different temperature,and their mutual effect on the (a) styrene conversion (b) Interaction Graph (c) Predicted vs. actual

171

Figure 4.37 Response surface plot showing the effect of oxidant and different temperature,and their mutual effect on the (a) benzaldehyde selectivity (b) Interaction Graph (c) Predicted vs. actual

172

Figure 4.38 Response surface plot showing the effect of oxidant and different temperature,and their mutual effect on the (a) styrene oxide selectivity (b) Interaction Graph (c) Predicted Vs. ACtual

173

Figure 4.39 FTIR spectra of polystyrene 174

Figure 4.40 Pore size distribution of unsupported membrane 177 Figure 4.41 Adsorption/desorption isotherms of unsupported sample 178 Figure 4.42 EDX analysis of membrane prepared (a) VxOy/CNT (b) VxOy-

NTs (c) VxOy/-Al2O nano particle

183 Figure 4.43 XRD patterns of unsupported membrane with different condition of

preparation

185

xx

Figure 4.44 Second order model for calculation reaction rate constants of the forward reactions at different temperature for oxidation of styrene with H2O2 using H2O2/styrene molar ratio (M=2 ) with catalyst (VxOy- Al2O3)

187

Figure 4.45 Second order model for calculation reaction rate constants of the forward reactions at different temperature for oxidation of styrene with H2O2using H2O2/styrene molar ratio (M =2) and catalyst (VxOy

-CNT)

188

Figure 4.46 Second order model for calculation reaction rate constants of the forward reactions at different temperature for oxidation of styrene with H2O2using H2O2/styrene molar ratio (M =2) and catalyst (VxOy

-CNT /-Al2O3)

188

xxi

LIST OF PLATES

page

Plate 3.1 Sintering temperature profile for support 63

Plate 3.2 photograph of liquid permeation membrane test rig 67

Plate 3.3 Assembling the reactor (a) membrane cell and suitable O-ring(b) fixing of cell inside reactor(c) Insert the mesh ring back into the first reactor(d) insert the O-rings into their prescribed grooves(e) Insert the screws and fasten them(f) Install both the ceramic heater

72

Plate 4.1 SEM of VxOy-NTs at different hydrothermal synthesis time and ultrasonic condition (a) with- out ultrasonic condition and 5 days hydrothermal treatment (b)15 min ultrasonic and 5 days hydrothermal treatment (c,d) 30 min ultrasonic 5 days hydrothermal tratment (e) 30 min ultrasonic 2 days hydrothermal treatment (f) 30 min ultrasonic and 3 days hydrothermal treatment

89

Plate 4.2 TEM images of VxOy-NTs at different hydrothermal treatment synthesis time and ultrasonic condition (a) 2 days hydrothermal treatment (b) 3 days hydrothermal treatment (c) without ultrasonic 5 days hydrothermal treatment (d) 30min ultrasonic 5days hydrothermal treatment (e,f) multi-walled nanotubes produced from 5 days hydrothermal treatment with 30 min ultrasonic treatment

91

Plate 4.3 SEM images of VxOy NTs after reaction (a),(b)VxOy-NTs after 15 min reaction with mixture of H2O2+Styrene(c),(d) VxOy- NTs after 60 min reaction with Styrene

96

Plate 4.4 7(0 LPDJHV RI -Al2O3 impregnated with V2O5 (a) bulk of alumina nano particle ( b) separate nano particle catalyst

98

Plate 4.5 TEM images of (a) MWCNTs (b,c) acid-treated MWCNTs, (d,e) MWCNTs coated with VxOy,(f) VxOy-&17VFRDWHGZLWK-Al2O3

sol

134

Plate 4.6 Effect of ultrasonic treatment on coating of membrane (a) surface of support before treatment (b) surface of support after treatment (c) coating with AlOOH + PVA without any support treatment (d) coating with AlOOH + PVA+PEG without any treatment on support(e) coating with AlOOH + PVA after support treatment (f) coating with AlOOH + PVA+PEG after support treatment

146

xxii

Plate 4.7 TEM images of (a)unsupported membrane PVA add to sol after peptization (b)unsupported membrane PVA add to sol before peptization (c)unsupported membrane PVA+PEG add to sol before peptization

152

Plate 4.8 SEM images of (a) membrane, PVA add to sol after peptization (b) membrane, PVA add to sol before peptization (c) membrane, PVA+PEG add to sol before Peptization

154

Plate 4.9 SEM images of (a) membrane with 0.01mol/l alumina sol concentration (b) membrane with 0.08 mol/l alumina sol concentration(c) membrane with 1wt% binder(d) membrane with 4wt% binder

156

Plate 4.10 Thickness of top layer with different of coating time (a) 6 sec (b)12 sec and(c)18 sec

160

Plate 4.11 surface of catalytic membrane (a) after reaction with temperature more than 65&EXVHGPHPEUDQHDIWHUFDOFLQDWLRQVDW& 174 Plate 4.12 SEM surface and cross section of (a,b)membrane with VxOy-

CNTs (c,d) VxOy-NTs(e,f) VxOy-Al2O3nano particles

180

Plate 4.13 AFM images of surface of membrane prepared with (a) VxOy- CNTs (b) VxOy-NTs(c) VxOy-Al2O3nano particles take at PP

182

xxiii

LIST OF ABBREVIATIONS

D^-Al2O3 Alfa-alumina

-Al2O3 Gama-alumina

AFM Atomic Force Microscopic

ANOVA Analysis of variance

AlOOH Boehmite

AAO Porous anodic aluminum oxide

BET Nitrogen adsorption analysis

Bza Benzaldehyde

Bzac Benzoic acid

CNT Carbon nanotubes

CMR Catalytic membrane reactor

CVD Chemical vapor deposition

CNT/VxOy Carbon nanotube/vanadium oxide

DOE Design of experiment

EDX Energy Dispersive X-ray

FT-IR Fourier transform infrared spectroscopy

GC Gas chromatography

GC-MS Gas chromatography-mass spectroscopy H4NO3V Ammouium metavanadate

h Hours

min Minutes

MWCNTs Multi wall carbon nano tube

NT Nano tube

xxiv

PEG Polyethylene glycol

PVA Polyvinyl alcohol

Pha Phenylacetaldehyde

PPM Porous polymeric material

PVP Polyvinylpyrrolidone

PEI Polyethylenimine

Phed 1-phenylethane-1,2-diol

RSM Response surface methodology

SO Styrene oxide

SEM Scanning electron microscopy

TEM Transmission electron microscopy

TIVO Triisopropoxy vanadium

TGA Thermal gravimetric analysis

TEOS Tetraethoxysilane

TMOS Tetramethoxysilane

VOC Volatile organic carbon

TS-1 Titanium silicalite-1

XxOy Vanadium oxide (VO,V2O3,V2O5,V6O13) VxOy-Al2O3 Vanadium oxide impregnated -Al2O3

VxOy-NTS Vanadium oxide nanotubes VxOy/ CNT Vanadium oxide impregnated CNT

XRD X-ray Diffractometer

1D One-dimensional

xxv

LIST OF SYMBOLS

Symbols Descriptions Unit

rst rate of reaction mol/(L h)

IJ time h

A pre exponential factor

t time h

T absolute temperature K

S catalyst surface area m²

X_st styrene conversion to benzaldehyde Xst0 initial styrene conversion to benzaldehyde Xoxidant0 initial oxidant conversion to benzaldehyde M molar ratio of oxidant (H2O2) and styrene

k0 specific rate of reaction L/(mol h)

ks overall rate constant L/(mol h)

Coxidant0 initial oxidant (H2O2) concentration mol/L

Cst0 initial styrene concentration mol/L

EA activation energy cal/(mol K)

e Error

l Thickness of coating m

p product

R² Coefficient of determination

E Activation energy cal/mol R gas constant

Vpore pore volume m³ Vmem membrane volume m³

membrane porosity

F Solution Flux lm^-2h^-1

xxvi

MEMBRAN PEMANGKIN KOMPOSIT NANO VANADIUM OKSIDA SECARA MATRIKS BERCAMPUR UNTUK PENGOKSIDAAN STIRENA

ABSTRAK

Pengoksidaan sebatian olefin kepada epoksida atau oksida setara dengan menggunakan pemangkin logam oksida adalah satu langkah penting dalam pembuatan bahan kimia gred tinggi dan farmaseutikal dalam kuantiti yang besar.

Penggunaan vanadium oksida sebagai pemangkin logam oksida mendapat perhatian khusus disebabkan keadaan pengoksidaannya dari +2 hingga +5 dan kepelbagaian sifat kimia stereo dengan julat koordinasi dari empat hingga lapan; menjadikannya logam oksida yang menarik dalam pengoksidaan olefin.

Struktur nano tersusun logam oksida seperti pertumbuhan logam oksida pada dinding tiub karbon nano (CNTs) adalah menarik dalam pemangkinan disebabkan luas permukaan tentu yang tinggi dan sifat-sifat uniknya. CNT terbenam vanadium oksida dijangka menghasilkan bahan komposit baru dengan sifat-sifat fizikal dan ciri-ciri pemangkin yang dipertingkatkan.

Kememilihan tertingkat dalam tindak balas pengoksidaan boleh dicapai melalui pemangkin matriks bercampur dan reka bentuk reaktor. Penggunaan reaktor membran bermangkin memudahkan persentuhan bahan-bahan tindak balas dalam fasa cecair. Dalam kerja ini, satu lapisan nipis membran bermangkin VxOy-

&17$O²O3(~ Ptelah disediakan dan dipantau dalam keupayaan pengoksidaan stirena. Logam oksida nanokomposit VxOy/CNT disediakan dengan kaedah pembentungan basah, kemudiannya diurai dalam sol AlOOH. Selepas itu, ia disalut pada permukaan penyokong menggunakan kaedah sol-gel. Faktor-faktor yang mempengaruhi penyediaan salutan tanpa kecacatan seperti jenis pengikat, masa untuk menambah pengikat ke dalam sol dan rawatan permukaan penyokong telah

xxvii

dikaji. Membran tiada penyokong dan berpenyokong telah dicirikan dengan menggunakan teknik-teknik seperti SEM, TEM , FTIR, TGA, AFM, XRD dan penjerapan-penyahjerapan nitrogen.

Reka bentuk eksperimen (DOE) digunakan untuk mengoptimumkan pengoksidaan stirena dan kememilihan benzaldehid. Bagi membran pemangkin, keadaan pengoksidaan optimum boleh dicapai pada suhu tindak balas C, perbezaan tekanan separa 1.5 bar, nisbah mol H2O2/stirena 1.5:1 dan beban pemangkin 30%; di mana penukaran maksimum stirena 25.6% dan kememilihan benzaldehid 84.9%. Manakala keadaan optimum dengan pemangkin nanokomposit VxOy/-Al2O3dicapai pada suhu tindak balas 62.74&GDODPWHPSRK5.15 j, nisbah mol H2O2/stirena 2.7:1 dan beban pemangkin 0.34 g; di mana penukaran maksimum stirena 68.23% dan kememilihan benzaldehid setinggi 57.32%.

Ia adalah jelas bahawa pengoksidaan stirena boleh diperbaiki dengan menggunakan pemangkin komposit berasaskan vanadium oksida dengan menggunakan H2O2 kering sebagai agen pengoksidaan. Jenis pemangkin, struktur pemangkin dan keadaan tindak balas mempunyai pengaruh yang kuat ke atas pengoksidaan stirena. Tiub nano vanadium oksida yang disintesis melalui kaedah hidroterma mempunyai prestasi yang rendah akibat kemusnahan pengoksidaan.

Walau bagaimanapun, kelemahan ini boleh diatasi dengan menggunakan CNTs sebagai acuan. Produk utama pengoksidaan stirena adalah benzaldehid. Dengan menggunakan VxOy-CNT/-Al2O3 sebagai membran pemangkin, penukaran stirena berkurangan tetapi kememilihan benzaldehid meningkat berbanding dengan serbuk nanokomposit disebabkan oleh pencegahan tindak balas sampingan. Membran boleh menghalang tindak balas sampingan dengan mengasingkan produk dari bertindak balas selanjutnya dengan H2O2.

xxviii

MIXED MATRIX VANADIUM OXIDE CATALYTIC NANOCOMPOSITE MEMBRANE FOR STYRENE OXIDATION

ABSTRACT

The oxidation of olefin compounds to equivalent epoxides or oxides using metal oxides as catalyst is an important step in the manufacturing of large quantities of fine and pharmaceutical grade chemicals. The chemical property of vanadium as metal in its oxide state is of interest particularly in application during oleQ e oxidation. Well-defined metal oxide nanostructures on the walls of carbon nanotubes (CNTs) are attractive as catalyst in the oxidation reaction because of their high specific surface area and formation of novel composite material with enhanced physical and catalytic properties. In this work, a thin layer of VxOy-&17-Al2O3

FDWDO\WLFPHPEUDQHDERXWPZDVSUHSDUHGDQGREVHUYHGIRULWVVW\UHQHR[LGDWLRQ

capability. Metal oxide nanocomposite VxOy/CNT was prepared by wet impregnation method dispersed in sol, AlOOH. It was then coated on the surface of support through sol-gel technique. Factors that affect the preparation of defect free coating such as type of binder, time for the addition of binder to sol and surface treatment of the support, were investigated. The physical and chemical properties of the unsupported and supported membrane were characterized using different techniques such as SEM, TEM, FTIR, TGA, AFM, XRD and nitrogen adsorption- desorption.

Design of experiments (DOE) was applied to optimize the styrene oxidation conversion and benzaldehyde selectivity using the prepared membrane and nanocomposite catalyst. The optimal oxidation conditions were achieved at reaction WHPSHUDWXUH RI & SDUWLDO SUHVVXUH difference of 1.5 bar, molar ratio of H2O2/styrene of 1.5:1 and loading of 30 wt. % catalyst on membrane. These

xxix

conditions resulted in maximum styrene conversion of 25.6% and benzaldehyde selectivity of 84.9%. In the case of nanocomposite catalyst (VxOy/-Al2O3), optimum oxidation conditions were reaction temperature of 62.74 & WLPH RI Kmolar ratio of H2O2/styrene of 2.7:1 and catalyst loading of 0.34 g for maximal styrene conversion of 68.23% and maximum benzaldehyde selectivity of 57.32%.

The results showed that styrene oxidation was improved using vanadium oxide-based composite catalysts with anhydrous H2O2 as an oxidant. Vanadium oxide nanotube, synthesized through hydrothermal technique also has relatively poor performance due to the oxidative degradation of template. However, using CNTs as a template, the shortcoming of vanadium oxide nanotube was overcome. Thus, using VxOy-CNT/-Al2O3 as catalytic membrane decreased styrene conversion and increased benzaldehyde selectivity (main product of styrene oxidation). This showed effectiveness of membrane to prevent side reaction, and isolate the product from further reaction with H2O2.

1 CHAPTER 1 1. INTRODUCTION

1.1 Membrane Technology

Today, membrane processes are at the centre of resolve and settlement of many important industrial problems, such as water treatment and desalination, food and beverages, pharmaceutical and medical applications, gas separation, full cell, biochemical reactions and catalytic reactions (Mulder, 1996).

Membranes are barriers, permeable to one or more species in a mixture, which separate two distinct zones and create a driving force for components to move from one zone to the other. Even though systematic studies of membrane phenomena may be traced as early as 18^thcentury, but not much of development was found until 20^th century. The first commercially significant gas separation membranes were introduced only in 1979, but within 10 years a wide range of different types of gas separation membranes have been developed (Noble and Stern, 1995). Moreover, membranes found their first considerable application in the filtration of drinking water at the end of World War II. By 1960, the fundamentals of modern membrane science had been developed, however membranes were used in only a few laboratories and small, specialized industrial applications (Mark, 2004). About thirty years ago, membrane filtration was not economically feasible, because of the disability of precisely controlling the pore size and pore morphology and also the specific materials that were used to produce the membrane, were restricted, but now this problem is settled and membrane technology has been further developed.

The two most important, and often the most expensive chemical process are usually the chemical reaction and the separation of the product stream. Both the

2

process could be improved by the combination of these two operations into a single unit operation, leading to potential savings in energy and reactant consumption and reduced by-product formation (Strathmann, 2001). One promising way to accomplish this combination is the use of membrane separation and catalytic reaction together in a multifunctional reactor.

The idea of catalytic membrane was first suggested by Sun in 1987 (Sun and Khang, 2002) thereafter, Burggraf (Burggraaf, 1989) indicated that the modification of -alumina membrane as catalytic membrane is possible. Over the past two decades, interest in membranes as a functional component of a reactor has significantly increased (Aran et al., 2011; Caro et al., 2010; Felice et al., 2010; Julbe HWDO.ORVH5RGULJXH]HWDO7DNHKLUDHWDO. Catalytic membrane reactor (CMR) is currently a challenging research subject in the field of membrane science and catalytic reaction engineering. Catalytic membrane reactor is one of the different membrane reactor configurations that are interesting because it could concurrently carry out two major operations: reaction and separation (Hsieh, 1991).

1.2 Application of Catalyst to Homogeneous and Heterogeneous Reactions Catalysts are used in approximately 90% of chemical, petrochemical and material manufacturing, so it is necessary to produce catalytic materials that are efficient, reusable over many cycles, and suitable for a wide range of processes (Maurya et al., 2011a). Catalysts do not change the thermodynamics of reaction, but slightly alter the kinetics of the reaction by providing an alternative pathway between the initial and final state of chemical reaction.

3

As shown in Figure 1.1, energy barrier is lowered with the introduction of catalyst in the process resulting in a more complex pathway of reactions. Therefore, catalyst increases the reaction rate constant, k by decreasing the activation energy, E as given in the Arrhenius relationship k= Ae^-E/RT.

Figure1.1:Effect of catalyst on a chemical reaction

Catalysts are commonly classified as enzymatic, homogeneous and heterogeneous (Swiegers, 2008). Homogeneous catalysts are located in the same phase as the reactants, and involve the reactions taking place on a single active site.

They are often comprised of single molecule. On the other hand, heterogeneous catalysts are solids, generally in the form of supported transition metals, which catalyze reactions of molecules in liquid or gaseous phases. As a result, heterogeneous catalysis takes place at an interface between a gas or liquid and a solid surface, where many different types of active sites like single atom or a group of atoms may be involved in the reaction. This might lead to a large number of products in spite of the fact that only one group might be desirable.

4 1.3 Nanoparticles Catalysis

The discovery of carbon nanotube (CNT) (Iijima, 1991) led to determining a unique structure for novel products, in addition to their physical properties and interesting shapes. There are several reports published on comprehensive development in synthetic routes and structure of nano-scaled materials and application of nano materials since 1991 (Ajayan, 1999; Cevc and Vierl, 2009).The synthesis of different kinds of inorganic nanomaterials was the motivation for chemists, physicists and material researchers to focus their attention toward the design of a variety of tubular or other types of structures containing nanoparticles (Jana et al., 2001; Shenton et al., 1999; Tremel, 1999).

The functionality of a heterogeneous catalyst depends critically on its structure over a length scales (Goodwin, 2004).Metal nanoparticles were likely observed over 2500 years ago (Daniel and Astruc, 2003). However, in the last 20 years, the study of nanoparticles has increased dramatically due to their unique size and properties. The interest for study and using nanoscale material is especially apparent in the field of catalysis because metal and metal oxide nanoparticles often show increases activity compared with their bulk metal counterparts (Okatsu et al., 2009; Polshettiwar et al., 2009). In heterogeneous catalysts, the favorable form of catalysts, are typically been stabilized or immobilized on different type of supporting material that will be discussed in detail as follow (Ishida et al., 2008).

1.4 Common Nanoparticle Supports

The efficient catalyst supports must have high surface areas to facilitate loading of the catalyst. In this section we discuss the various types of nanoparticle supports.

5 1.4.1 Metal Oxide Supports

Metal oxides are one of the most extensively employed catalyst supports since they offer high thermal and chemical stability. While they are in the form of zeolites, they as well have a well-defined pore structure and a high surface area. The surfaces of metal oxides can be simply functionalized, to simplify the procedure for catalyst deposition. The most common metal oxides used as catalyst supports are silica, alumina, titania, ceria, and zirconia.

Silica is normally used in catalytic reactions that require mild temperatures,

< 300 ¡C, because at higher temperatures, it is less stable and can produce volatile hydroxides. Alumina offers higher thermal and mechanical stability than silica, thus mainly used as metal oxide support. Alumina exists in a number of forms, but the two most common are -alumina and -alumina. -Alumina is a highly porous, amorphous material that offers surface areas as high as 300 m²/g and pore sizes as small as 5 nm, on the other hand,-alumina is a nonporous, crystalline solid that has a relatively low surface area (3-5 m²/g). -alumina is highly stable even at temperatures as high as 1200 ^oC (Chorkendorff, 2003).

Silica and alumina are characteristically classified as inert materials, compared to other metal oxide supports such as titania, ceria, and zirconia which show reactive properties in certain reactions (Boaro, 2009; Han, 2009).

1.4.2 Carbonaceous Supports

Carbon materials offer a multiplicity of advantages as catalyst supports. Some carbon supports have surface areas as high as 1500 m²/g and pore sizes less than 1 nm, but graphitic carbon has a moderately low surface area. Much of the present research with carbon supported catalysts focuses on developing novel carbon

6

materials such as carbon nanotubes as catalyst supports (Satishkumar et al., 2000) Carbon nanotubes are unique substrates for catalyst immobilization because of their high surface area, unique physical properties and morphology, and high electrical conductivity. Additionally, their small size and hollow geometry facilitates the formation of small nanoparticles (typically 1 - 4 nm), which is ideal for catalysis.

1.4.3 Polymer Supports

As an option to conventional metal oxides and carbon, polymeric materials are attractive as nanoparticle supports because of their adaptability. Many polymeric materials contain heteroatom that can form complexes with metal nanoparticles, and the flexible structure of the polymers often makes them principally effective at stabilizing metal nanoparticles and preventing aggregation. Also, the variety of functional groups available in polymeric materials is almost unlimited. Some of the most common types of polymers used for nanoparticle immobilization include water soluble polyelectrolytes such as polyvinylpyrrolidone (PVP) and polyethylenimine (PEI) (Sun and Wang, 2003; Tsunoyama et al., 2008; Tsunoyama et al., 2005), polymeric microspheres (Liu, 2006), and ion exchange resins (Ishida, 2007; Shi, 2005 ).

1.4.4 Porous membranes as catalyst supports

As mentioned above, high surface areas result from a highly porous structure with comparatively small pore sizes, however small pores often exhibit high mass transport resistances during the catalytic reaction. Porous membranes, which can consist of virtually any ceramic or polymer material, offer a support configuration that provide several advantages over the conventional catalyst supports (Armor, 1995). One of the motivations for using membranes as catalyst supports is the ability

7

to catalyze reactions and perform separations at the same time. Also, membrane reactions can run continuously because the catalyst does not need to be separated from the reaction products, unlike batch and stirred tank reactors. The internal pores of the membrane also provide a large surface area that permits a high loading of the catalytic material.

1.5 Membranes Reactors

Common membrane materials can be separated into two main classes, polymeric (organic) and inorganic. The main limitations of polymeric membranes are their low operating temperatures and chemical stability. Inorganic membranes can operate at higher temperatures and are typically more chemically resistant. Table 1.1 lists a number of membrane materials (polymeric and inorganic) along with their maximum operating temperature and the suitable pH range. Inorganic membranes are generally more resistant to organic solvents and corrosive chemicals than the organic membranes, and are more suitable for use in most reaction systems due to their higher thermal and chemical stability.

The application of membranes in chemical reactors is motivated mostly by the synergy effect that is created by the preferential permeation of products (or reactants), leading to higher conversion and/or selectivity, and a potentially reduced downstream separation load. There has been an intense, worldwide effort on membrane reactors research VLQFHWKHsVDQGWKHVHHIIRUWVKDYHEHHQVXPPDUL]HG in a number of recent review articles (Andric et al., 2010; Basile et al., 2006; Fong et al., 2008; Giorno and Drioli, 2000; Li et al., 2009; Mozia, 2009; Shu et al., 1991;

Westermann and Melin, 2009).

8

Table 1.1:Operating limits and restrictions for various membrane materials (Hsieh, 1996)

Material Maximum Operating

Temperature (^oC)

pH Range

Cellulose acetate 50 3-7

Aromatic polyamides 60-80 3-11

Fluorocarbon polymers 130-150 1-14

Polyimides 140 2-8

Nylons 150-180 *na

Polycarbonate 60-70 *na

Polyvinyl chloride 120-140 *na

Alumina >900 0-14

Glass 700 1-9

Zirconia 400 1-14

Silver 370 1-14

Stainless steel(316) >400 4-11

*na= not available

Depending on the configuration of membrane reactors, they may provide a number of other advantages as well. Catalytic membranes are typically divided into four categories: retainment of homogeneous catalysts, extractors, distributors, and contactors, which are discussed below in detail.

1.5.1 Retainment of Homogeneous Catalysts

Thermal recovery of homogeneous catalysts is regularly uneconomical due to low concentrations and can lead to deactivation of the catalyst. Retention of a homogeneous catalyst or precious smaller ligands, allowing the permeation of the residual reaction mixture could be achieved using membrane reactor as shown in Figure1.2. Membrane bioreactor for example, is particularly popular in bio catalytic applications (Giorno and Drioli, 2000; Westermann and Melin, 2009) and can be either realized by immobilizing the catalyst on the membrane surface or by a membrane filtration with a dissolved or dispersed catalyst.

9

Figure 1.2:Retainment of homogeneous catalysts (Seidel-Morgenstern, 2010a)

1.5.2 Membrane Reactors for the Preferential Removal of a Species (Membrane Extractors)

The function of a membrane extractor is to selectively remove a product from the reaction mixture.In the late sV researchers began to exploit the excellent separation capabilities of membranes for catalytic applications. The earliest work in this area employed membrane extractors to remove one or more of the products from the reaction to shift the equilibrium and increase its conversion. If the reaction is equilibrium-limited, the decreased activity of the species being removed will permit further conversion to occur, beyond that which would be possible if no species were removed. When reactions are limited by equilibrium, the membrane can continuously and selectively remove the product to shift the equilibrium.

Many of the initial studies with membrane extractors involved beneficial hydrogen removal in reactions such as hydrocarbon dehydrogenation (Collins et al., 1996; Hšllein et al., 2001), water gas shift (Criscuoli et al., 2001), and methane steam reforming reactions (Shu et al., 1994). Figure 1.3 shows a membrane reactor that preferentially removes a species from an equilibrium limited reaction and hence improving the reaction conversion.

Reactants

Homogeneous catalyst

Products Ji&DWDO\VW

10

Figure 1.3:Extractor Membrane(Seidel-Morgenstern, 2010a)

1.5.3 Membrane Reactor for Distributed Addition of a Reactant (Membrane Distributors)

The idea of a distributed reactant feed is useful to systems with two competing reactions. A typical example is the partial oxidation of a hydrocarbon.

Distributor membrane reactors present controlled addition of a reactant to the reaction mixture to limit side reactions and provide higher selectivity for the desired product. This may happen by simply controlling the amount of a reactant that is introduced on one side of the membrane, or by selectively allowing one component from a mixture to pass through the membrane and undergo reaction.

Figure 1.4 demonstrates the reaction of two components, A and B, to yield a product, C without the continued formation of D. As component B permeates through the membrane, it reacts with component A on the other side of the membrane to form product C. The reaction of B creates a concentration gradient within the membrane, which is the driving force for the continued permeation of B and subsequent reaction with A. Because B reacts with A immediately upon permeating through the membrane, very little side reactions occur to produce D. This principle is complicated by the change in residence time behavior of the reactants as one of them is fed gradually (Klose et al., 2003).The most common example of this type of reaction is the partial oxidation of hydrocarbons (Kšlsch et al., 2002). In these reactions, O2 reacts more easily with the partially oxidized species than the

11

starting material. By controlling the addition of O2 (B) to the hydrocarbon stream (A), complete oxidation can be limited.

Figure 1.4:Membrane reactors for distributed addition of a reactant (Seidel-Morgenstern, 2010) 2010)

1.5.4 Membrane Reactors for Control of Reactant Contact (Membrane Contactors)

The newest form of catalytic membrane reactor is the membrane contactor, where the membrane acts as a support that brings the reactants into contact with the catalyst. In the case of extractors and distributors, the membrane is often catalytically inert and simply acts as a support with a fixed bed of catalyst on one side of the membrane. In that case, the membrane is solely responsible for the separation step, while the fixed catalyst bed is responsible for the reaction. Membrane contactors typically have catalyst deposited within the membrane pores, making them catalytically active, but they usually do not perform a separation function. Although, they do not fulfil the requirements for catalytic membrane reactors in the traditional sense (combining separation and reaction), most researchers in the field still characterize them as a form of catalytic membrane reactor. Membrane contactors are further classified into two categories in which the membrane acts as an interface between multiple phases (Figure 1.5 a) or the membrane is utilized for dead-end flow

12

where all the reactants pass through the catalytically active membrane pores (Figure 1.5 b) (Miachon et al., 2003).

Figur 1.5:Membrane reactors for control of reactant contact (Seidel-Morgenstern, 2010a)

1.5.4 (a) Interfacial Contactors

Interfacial contactors, which are also known as catalytic diffusers, exploit the membrane as a catalyst support to facilitate contact between reactants that are in multiple phases. The membrane could cause enhanced contact between the catalyst and the reactants. The common application for interfacial contactors is gas/liquid contacting for reactions such as hydrogenations or oxidations.

A gas/liquid contactor typically has the gas on the side of the membrane that has larger pores and the liquid on the side that contains more of the catalyst, which is usually the skin layer. The gas and liquid solutes then diffuse to the catalyst surface and react. Flow of the liquid and gas facilitates transport of reactants and removal of products from the reaction zone. This configuration was successfully applied to oxidize organic acids to CO2and H2O (Miachon et al., 2003) and short chain alkanes to oxygenates (Espro et al., 2001). A number of studies had been carried out for hydrogenation of unsaturated substrates including cinnemaldehyde (Pan et al., 2000), methylenecyclohexane and sunflower seed oil (Bottino et al., 2002) with excellent success, in some cases achieving high selectivity for the desired product. There has

(a) (b)

13

also been significant interest in using interfacial contactors for removal of nitrates from drinking water by hydrogenation (Reif and Dittmeyer, 2003).

The interfacial contactor configuration is also beneficial for forming contact between two liquid phases. The reactants diffuse into the pores from each side of the membrane and come into contact at the catalyst surface.

1.5.4(b) Flow-Through Contactors

Membranes operated as flow-through contactors act similarly to conventional fixed bed reactors. In traditional fixed bed reactors, the reactants need to diffuse into the pores of the support to react with the catalyst, but in flow through membrane contactors, convective mass transport rapidly brings the reactants to the active surface of the immobilized catalyst (Figure 1.6). If the pores are adequately small, radial diffusion will not limit reactions even at high flow rates where the contact time is very short. In this way, the pores of the membrane act as microreactors in which the reaction conversion can be controlled by simply adjusting the flow rate (Westermann and Melin, 2009). Flow-through contactors also have the advantage of constantly removing the products from the reaction zone. As a result, there will be less competition between products and reactants for active sites on the catalyst and a decreased possibility for the products to undergo side reactions or poison the catalyst surface.

The flow-through contactor configuration allows fine control over the residence time of reactants inside the membrane, which directly affects conversion.

Even a single pass through the membrane yields nearly complete conversion in reactions such as volatile organic carbon (VOC) combustion or photocatalytic oxidation of organic compounds. However, a majority of previous experiments with

14

flow-through contactors passed the reactant mixture through the membrane multiple times to obtain complete conversion. Many of these studies focused on gas-phase hydrogenation or oxidation reactions (Kormann et al., 2004).

Fgure 1.6:Schematic diagram showing the flow of reactants through a membrane operated as a (a) flow-through contactor, and around a catalyst bead from a (b) traditional fixed bed reactor.

The flow of gaseous mixture through the membrane very rapidly allows very short contact time, which results in partial hydrogenation of compounds such as acetylene, propyne, butadiene, and hexadiene (Groschel et al., 2005; Lambert and Gonzalez, 1999; Lange et al., 1998; Pelzer et al., 2003). The flow-through configuration is also beneficial for gas/liquid reactions where a limited amount of gas is able to dissolve in the liquid phase. Thus with each pass, the liquid solution can again be re-saturated with the gas. Schmidt and co-workers found that this strategy afforded high selectivity in the partial hydrogenation of a number of unsaturated organic compounds including cyclo-octadiene, 1-octyne, phenyl acetylene, and geraniol (Schmidt et al., 2005; Schmidt et al., 2008). Others had similar success with partial hydrogenation of sunflower oil (Bengtson and Fritsch, 2006; Fritsch and Bengtson, 2006; Schmidt et al., 2008) and a-methylstyrene (Purnama et al., 2006) and hydrogenation of nitrate for water denitrification (Ilinich et al., 2003; Reif and

(a) (b)

15

Dittmeyer, 2003). The overall success of membrane reactors in restricting the extent of reaction could make them quite valuable in a number of applications.

1.6 Benzaldehyde Synthesis and Application

Benzaldehyde C6H5CHO, is the simplest and the most industrially useful member of the family of aromatic aldehydes. Benzaldehyde exists in nature, primarily in combined forms such as glycoside in almond, apricot, cherry, and peach seeds. The characteristic benzaldehyde odor from oil of bitter almond occurs because of trace amounts of free benzadehyde formed by hydrolysis of the glycoside amygdalin. Amygdalin was UVWLVRODWHGLQIURPWKHVHHGVRIWKHELWWHUDOPRQG

The only industrially important processes for the manufacturing of synthetic benzaldehyde involve the hydrolysis of benzalchloride and the air oxidation of toluene. The hydrolysis of benzalchloride, which is produced by the side-chain chlorination of toluene, is the older of the two processes. It is not utilized in the United States in Europe, India, and China. Other processes, including the oxidation of benzylalcohol, the reduction of benzoylchloride, and the reaction of carbon monoxide and benzene, have been utilized in the past, but they no longer have any industrial application(kirk-Othmer, 1999).

The air oxidation of toluene iV WKH VRXUFH RI WKH PDMRULW\ RI WKH ZRUOGsV synthetic benzaldehyde. Both vapor and liquid-phase air oxidation processes have been used. In the vapor phase process, a mixture of air and toluene vapor is passed over a catalyst consisting of the oxides of uranium, molybdenum, or related metals.

High temperatures and short contact times are essential to maximize yields. Small amounts of copper oxide may be added to the catalyst mixture to reduce formation of by-product maleic anhydride. Conversion per pass is reported to be low, 10w20%,

16

with equally low yields, 30-50%. The vapor-phase oxidation of toluene was the dominant process in the 1950s and early 1960s, but is now of little industrial importance, thus paving way for the liquid-phase process. In the liquid phase process, both benzaldehyde and benzoic acid are recovered. This process was introduced and developed in the late 1950s by the Dow Chemical Company, as a part of their toluene-to-phenol process, and by Snia Viscosa for their toluene-to- caprolactam process (kirk-Othmer, 1999). The benzaldehyde recovered from the liquid-phase air oxidation of toluene may be purified by either batch or continuous distillation.

1.7 Problem Statement

-Al2O3has been widely studied for preparation and modification of catalytic membranes using the solwgel method because these types of membranes can be used for oxidation reaction purposes (Alfonso et al., 2000; Bottino et al., 2005b).-Al2O3

PHPEUDQHVDUHW\SLFDOO\VXSSRUWHGRQPDFURSRURXV-Al2O3 tubes or disks in the sol-gel method; they are dip or spin coated with a boehmite (AlOOH) precursor to improve conversion or selectivity (Xia et al., 1996).

However, when AlOOH is applied, the surface on which -Al2O3 is catalytically embedded must be free of cracks after calcination. Therefore, choosing a suitable binder that can be burned off in the calcination step without negative side effects on the catalyst or membrane is important. Although catalytic membranes with different shapes, catalyst loadings and binders have been prepared and tested in commercial ceramic support structures (Alfonso et al., 1999; Alfonso et al., 2002;

Teixeira et al., 2011), to the best of our knowledge, no studies on the use of vanadium oxide (VxOy) supported on multi-walled carbon nanotubes (CNTs) or

17

vanadium oxide NTs for styrene oxidation in catalytic membranes have been investigated. The reason is because no suitable and effective catalyst has been developed for this purpose.

The large surface area 1315 m²/g for SWNTs, outstanding thermal conductivity (3000W/K m) and chemical and mechanical stabilities of CNTs, make them promising for the development of advanced composite materials that undergo catalytic oxidation reactions (Chen et al., 2007b). As a result of their amazing properties, CNTs have been introduced into many host materials, including polymers, metals and ceramics, to improve the overall properties of CNT composite systems (Sharma et al., 2010; Thostenson et al., 2001).

The growth of metal oxides on the walls of carbon nanotubes (CNTs) is a common technique for preparing composite materials (Neri et al., 2010). The use of vanadium oxide nano-structures as metal oxide catalysts is of particular importance because vanadium oxides can be used for the partial oxidation or dehydrogenation of alkanes to oleQV (Ledoux et al., 2001; Weckhuysen and Keller, 2003a; Xu et al., 2002). Therefore, blends of vanadium oxides and CNTs are expected to produce novel composite materials with enhanced chemical and physical properties (Chen et al., 2007b).

Embedding carbon nanotubes and vanadium oxides into -Al2O3 poses several critical challenges that must be overcome to capture the full potential of CNT composites. The uniform dispersion of CNTs in the host matrix material is ess